January 10, 2011

The Vizzini Leak

Man in Black:  All right. Where is the poison? The battle of wits has begun. It ends when you decide and we both drink, and find out who is right ... and who is dead.

Vizzini:  But it's so simple. All I have to do is divine from what I know of you: are you the sort of man who would put the poison into his own goblet or his enemy's? Now, a clever man would put the poison into his own goblet, because he would know that only a great fool would reach for what he was given. I am not a great fool, so I can clearly not choose the wine in front of you. But you must have known I was not a great fool, you would have counted on it, so I can clearly not choose the wine in front of me.

Man in Black:  You've made your decision then?

Vizzini:  Not remotely. Because iocaine comes from Australia, as everyone knows, and Australia is entirely peopled with criminals, and criminals are used to having people not trust them, as you are not trusted by me, so I can clearly not choose the wine in front of you.

Man in Black:  Truly, you have a dizzying intellect.

Vizzini:  Wait 'til I get going!

The Princess Bride

The poison-duel scene in The Princess Bride is a comedy classic in which the villain, Vizzini, matches wits with the hero, the "Man in Black". Vizzini works through elaborate rationalizations for why the poison may have been placed in each of the two wine goblets, hoping to catch a tell by the Man in Black as to which goblet is truly poisoned. Eventually, Vizzini manages to talk himself into and out of believing that the poison is in each of the two goblets. Finally, he resorts to a ruse to try to cheat his way to the right solution, only to discover ... well, watch the scene and find out!

I often think of the Princess Bride poison-duel when confronted with a poker player who "tanks" for a ridiculously long time during a hand. I can only imagine the inner dialogue must run something like this:

"He overbet the river, so he's trying to look strong, so he must be weak, so I have to call. Except he knows that I know that "strong means weak", so he must actually be strong, pretending to be weak, so he looks too strong, knowing I will think he's really weak. So, clearly I must fold. However, he's a young guy wearing an Ed Hardy hoodie, so he's a big bluffer, so I must call. But he's only shown down the nuts tonight, and he has a tattoo of a donkey sh*tting diamonds out his azz, so he must have a strong hand, so I must fold. But ..."

It's actually kind of humorous to see some of these deep-thought moments at the table, when it's pretty obvious to most of the table whether the bet represents a value bet by an almond broker, or yet another bully bluff by the table maniac. It's truly astonishing how often these deep thinkers make the wrong decision.

A psychological quirk that might be in play in these situations is the surprising fact that deep analysis often becomes over-analysis, and over-analysis is often no better than a simple wild azz guess in terms of results. The fault in our thinking process arises from attempting to factor into our decision far too many factors, leading us to put undue weight on peripheral factors, or to minimize the weight given to key factors:

This is known was a “weighting mistake,” and it’s a serious problem for conscious deliberation. When we try to analyze our alternatives, we tend to search for reasons to choose one team over another. The problem is that we’re not particularly good at figuring out whether or not these reasons are relevant. In other words, we’re rationalizing, which is quite different from being rational.

—Jonah Lehrer, "You Know More Than You Know" in The Frontal Cortex (Oct. 12, 2010).

Studies have shown that the best decisions by experts are reached not by an immediate "gut" reaction, nor by a deeply analytical thought process, but rather by "intermediate decisions", where an expert is given the problem to consider, and then briefly distracted by other matters before being asked for their conclusion. In that situation, the expert has sufficient opportunity to allow his brain to rationally process the decision in light of the expert's knowledge and experience, but prevents the brain from over-analyzing the situation and out-thinking itself.

The upshot of this research is that, while we should approach problems rationally, sometimes more thinking is poor thinking. Although our immediate gut reaction at the poker table might provide a valuable starting point, it pays to give some extra thought before making our final decision. But it pays equally well if we learn when to turn off the logic circuits, because ultimately we can rationalize almost any play we want to make if we give our brain time to focus on the factors supporting that decision.

As any Princess Bride fan knows, two of the classic blunders are:  "Never get involved in a land war in Asia", and "Never go against a Sicilian when death is on the line." When it comes to poker, let me add this blunder to the list:

"If you find yourself talking yourself into a call or a fold, you're an idiot if you listen to that fool."

"You have defeated my Scandi, and bested my Grinder ..."


  1. Inconceivable!

    (Can't believe I'm the first one to post that)

  2. Damn, Andy beat me too it. Outstanding analogy though and great breakdown of the topic.

  3. @ Andy: I don't think that word means what you think it means.

  4. I enjoy playing heads-up SNGs and for months was a barely break-even player. After analyzing my leaks, I found the biggest one to be just that... overanalyzing players that should never have been afforded that much thought. I gave most players too much credit for "trickiness" instead of realizing the majority were just bad players who could be bested by mere patience and discipline. By simply waiting for playable hands, not trying to bluff as much, I've raised my ROI to above 10% and have moved up two buyin levels that I can now beat regularly.

    Great topic and advice.

  5. Great post, thank you.

    I've seen other bloggers (it might have been the Grump) write that one of their biggest leaks is thinking they are being bluffed when they aren't. Sometimes an apple is just an apple (or something like that). Maybe that's another example of over-analyzing.